In cerebellum-like circuits, multiple sensory and motor streams are integrated by principal cells to compare expected and actual sensory information, a process that involves transforming input signals so that they can be effectively compared. The unipolar brush cell (UBC) is a glutamatergic interneuron in cerebellum-like systems that profoundly reshapes the duration and even polarity of input signals carried by mossy fibers. UBCs make contacts onto granule cells and other UBCs. Recent work in electric fish showed that transformation by UBCs is essential for generating images of corollary discharge necessary for synaptic plasticity. In mammals, UBCs are found in vestibular cerebellum and the cerebellum-like dorsal cochlear nucleus (DCN), yet the function of these mammalian neurons is not clear. Our work has shown that subtypes of UBCs differ dramatically in how they alter their glutamatergic mossy fiber signals, with ON UBCs responding to glutamate with prolonged excitation and OFF UBCs responding with inhibition or profoundly delayed excitation. How these ON/OFF signals are generated requires clarifying the synaptic mechanisms of UBCs, mechanism that may be unique in the brain. Knowing the function of these cells requires defining what signals are actually carried by the diverse mossy fiber inputs to UBC subtypes. Also unknown are the actual synaptic targets of each UBC subtype and how the UBCs alters the firing patterns of those targets. This proposal answers these major questions in the field in both vestibular and auditory regions using cutting edge anatomical and electrophysiological approaches in mice.